Designing AI-powered patient simulations for nursing students at UMass Boston
2025 - Product Design & User Research
OVERVIEW
Patient Proxy is a conversational AI simulation tool designed to help nursing students at UMass Boston practice patient communication in realistic clinical scenarios. Students can create patient profiles, interact with an AI-powered simulated patient through a chat interface, and receive feedback on their performance after completing a simulation. As the product designer, I designed the overall experience and conversational interface, while collaborating with an AI engineer who implemented the underlying LLM system.
View Live Demo
My Role
As the UX designed, I designed the overall Patient Proxy experience, shaping how nursing students interact with the conversational AI during simulation exercises. To ensure the design reflected real clinical training, I met with the Director of Nursing Simulations at UMass Boston to understand how simulations are conducted and how student performance is evaluated. I also conducted usability testing with nursing students at UMass Boston to gather feedback and refine the experience.
Duration
4 Months
THE PROBLEM
Nursing students need opportunities to practice patient communication before entering real clinical environments. Traditional simulation training can be limited by time, resources, and instructor availability. As a result, students often have fewer opportunities to practice realistic patient conversations before clinical placements.
THE SOLUTION
Patient Proxy is a conversational AI simulation platform that allows nursing students to practice patient communication in a safe and repeatable learning environment.
Interactive Patient Simulations
Custom Patient Profile
Simulation Feedback
UNDERSTANDING THE SIMULATION PROCCESS
To design an experience that reflected real clinical training, I held weekly syncs with the Director of Nursing Simulations at UMass Boston. During these meetings, she shared domain knowledge about how nursing simulations are conducted, including how students interact with mannequins, how scenarios are structured, and how instructors evaluate student performance.
These conversations helped me understand how students practice patient communication in simulation environments and informed how the digital experience should mirror the workflows used in UMass Boston’s nursing program.
DESIGNING THE SIMULATION EXPERIENCE
To mirror how simulations are conducted at UMass Boston, I designed the platform around a conversational interface that allows students to interact with a simulated patient in real time. The goal was to replicate the communication-focused nature of nursing simulations while keeping the interface simple enough for students to focus on the patient interaction rather than navigating complex controls.
Key design considerations included:
• maintaining a natural conversation flow
• keeping patient information easily accessible
• minimizing distractions during simulations
KEY DESIGN DECISCIONS
Designing for natural conversation
Nursing simulations focus heavily on communication between the student and patient. To support this, I designed the interaction around a chat interface that allows students to ask questions and gather symptoms in a conversational format.
Supporting context during simulations
During simulations, students often need access to patient information and scenario details. I designed navigation that allows students to quickly reference this information without interrupting the conversatio
PRODUCT WALKTHROUGH
Simulation Interface
Students interact with the simulated patient through a chat-based interface that mirrors natural clinical conversations. The layout prioritizes readability so students can focus on patient communication during the simulation.
Simulation Interface
Students interact with the simulated patient through a chat-based interface that mirrors natural clinical conversations. The layout prioritizes readability so students can focus on patient communication during the simulation.
Simulation Interface
Students interact with the simulated patient through a chat-based interface that mirrors natural clinical conversations. The layout prioritizes readability so students can focus on patient communication during the simulation.
TECHNICAL IMPLEMENTATION
The simulation uses a large language model to generate patient responses dynamically during conversations. The system processes student questions and produces responses based on the selected patient profile and scenario context.
I collaborated closely with an AI engineer, David Martinez, who implemented the technical infrastructure and built the application using Vercel. I designed the interface and simulation flows in Figma, and David translated those designs into a working web application that supports real-time simulated patient conversations and post-simuation feedback.
REFLECTION
Working on Patient Proxy strengthened my interest in designing tools that support healthcare education. Collaborating with an AI engineer also gave me insight into how design decisions translate into technical implementation. If the project continued, I would explore deeper feedback insights for students and ways to support instructors in tracking simulation performance over time. This experience reinforced my curiosity about how thoughtful design can improve healthcare learning environments and help prepare students for real-world patient interactions.